Mild Guillain-Barré Syndrome

Background  The unpredictability of the early course of Guillain-Barré syndrome (GBS) makes it difficult to determine which patients' conditions will worsen under observation. Most large randomized treatment trials for GBS have used an inability to walk as the enrollment criterion. Consequently, little is known about the treatment of those patients with milder degrees of affection.

Objectives  To determine the approximate frequency of mild GBS with the persistent ability to walk and to see if there were features that predicted that the illness would remain mild.

Setting  A registry of patients with GBS seen on the wards and in the neurology clinic from January 1,1992, to May 1, 2000, in a 400-bed community teaching hospital.

Patients  Twelve (4.7%) of 254 patients in our case series were able to walk throughout their illness. Eight had been treated with plasmapheresis or intravenous immunoglobulin; the others were observed without treatment.

Results  There was no age, sex, or seasonal preponderance in comparison with large case series that included cases of all severities. Nine of 12 patients had a preceding respiratory tract infection, 10 had paresthesias, 7 had prominent pain, and 9 had ataxia. Seven of 10 patients who were examined had normal cerebrospinal fluid protein levels. It took 8 days, on average, to reach the maximal degree of weakness. One additional treated patient, excluded from our case series, had mild weakness for the first 3 weeks and subsequently worsened with a relapsing course more typical of chronic inflammatory demyelinating polyneuropathy. Eleven patients demonstrated proximal, intermediate, or distal conduction block, and only 3 had a mild degree of denervation. There were no distinguishing clinical or electrophysiologic features between treated and untreated patients with mild GBS and, except for the mild degree of affection and the absence of substantial electromyographic changes of axonal disruption, there were no important differences between these mild cases as a group and patients who developed more severe GBS.

Conclusions  Cases of mild GBS reach a clinical nadir in a similar time to those with more severe disease. Treatment may be unnecessary in patients who are able to walk during the second week of illness, but observation until approximately the eighth day seems appropriate to be certain that the illness does not progress. In all likelihood there are mild cases of GBS that never come to the attention of a neurologist.

GUILLAIN-BARRÉ syndrome (GBS) is generally easy to identify by its typical clinical features and by the progression of weakness over several days. In most large randomized trials that have established the effectiveness of immune treatments for GBS, the enrollment criterion has been the patient's inability to walk (Medical Research Council grade 3 or higher).¹ The problem often arises regarding treatment for a patient with GBS who has not reached this threshold; for example, one who is still able to walk without assistance. Little is written about this group of patients and the unpredictability of the early clinical course of GBS makes it difficult to judge which patient's condition will worsen under observation. We determined the approximate frequency of this type of mild GBS, looked for features that predicted the persistence of this lesser degree of affection, and considered whether treatment might be avoided in these individuals.

Cases of mild GBS seen on the wards and in the neurology clinic from January 1, 1992, to May 1, 2000, were selected from the registry of St Elizabeth's Medical Center, Boston, Mass. Patients were chosen if they were able to walk without assistance (Medical Research Council grades 1 or 2) until they began to improve.¹ They otherwise had the typical features of bilateral limb weakness, distal paresthesias, mild sensory changes, some with, and others without, cranial nerve weakness. Patients with variants of GBS such as Fisher syndrome, pure or predominantly cranial nerve involvement, or purely sensory impairment were not considered. One patient with chronic inflammatory demyelinating polyneuropathy that became apparent later is considered separately. Each patient underwent electromyography (EMG); all but 2 had cerebrospinal fluid (CSF) studies performed at the early stage of the illness.

To summarize the extent of weakness, a motor scale was devised by adding the sum of 4 measurements of power derived from the proximal and distal parts of all 4 limbs on the Medical Research Council muscle power grading scale (each measurement ranging from 0, no movement, to 5, normal power). The summed motor score constituted a total of the following 4 averaged strength measurements: both arms proximally (deltoid, biceps, and triceps), both arms distally (wrist extensors and interossei), both legs proximally (iliopsoas, quadriceps, and hamstrings), and both legs distally (tibialis anterior and gastrocnemius). A normal motor examination resulted in a score of 20. We also recorded the average vital capacity for each day or the single measurement for the day.

Twelve (4.7%) of the 254 patients examined during the study period were able to walk throughout their entire course of GBS. A typical example was a 52-year-old man (patient 3) with an upper respiratory tract illness who, 2 weeks later, developed lower back pain and paresthesias in the hands and feet. Four days later he had diplopia and difficulty walking. Six days after onset he was at his weakest with a unilateral abducens palsy, mild symmetric proximal more than distal weakness in the arms and legs, universal areflexia, diminished vibration sensation at the toes, and a wide-based gait. The lowest vital capacity during his hospital course was 3.1 L. On the second day of illness, the CSF protein level was 490 g/L (49 mg/dL) and the nerve conduction studies showed proximal motor nerve conduction block, and absent or reduced sensory nerve action potentials. The needle EMG demonstrated no denervation. He completed a course of intravenous immunoglobulin and was given intravenous steroids for diffuse pains in the lower back, neck, shoulders, and hands. His strength was nearly normal after 3 weeks, he returned to work by 2½ months, but still had dysesthesias of his hands.

One additional patient appeared to have a mild form of GBS over 3 weeks but worsened 2 months later, requiring a wheelchair. She was believed to have chronic inflammatory demyelinating polyneuropathy based on later relapses and was excluded from this case series.

The clinical characteristics of the 12 patients with mild GBS, including the lowest motor score and the lowest vital capacity, are given in Table 1. Their ages ranged from 19 to 64 years. Nine patients had a preceding upper respiratory tract infection, but they did not cluster at any particular time of the year. All but 2 had paresthesias. Pain was a significant feature in 7 patients; 2 had lower back pain, but others reported periarticular pain or more diffuse pain involving not only the lower back, but also the neck, shoulders, hands, and calves. One patient developed proximal and distal dysesthesias in all 4 limbs. Another patient continued to have substantial pain even 6 months after the onset of the illness. Thse were lancinating pains that initially involved the proximal upper extremities but during the next 2 weeks became situated primarily in the patient's hands and were severe enough to require the administration of pain medications and epidural steroid injections.

The worst motor scores ranged from 14 to 19 (20 being full strength) and remained essentially unchanged or improved mildly during observation. Maximal weakness occurred within 1 week in 7 patients and within 14 days in all patients; the average time to nadir was 8 days. All were areflexic or had a few trace reflexes present. All but 3 had some degree of ataxic gait, generally minimal. Four of the 12 patients had cranial nerve palsies, mostly facial. None had bladder or autonomic involvement. The vital capacity was above 1.74 L in all patients, most exceeding 2.4 L. These measurements were stable during the hospital course. The CSF protein level was elevated in only 3 of the 10 patients in whom it was measured, but in some instances the normal protein level may have been related to the brief duration of illness at the time of sampling.

All of the patients had an EMG between the 2nd and 21st day of illness. Only the first EMG results are listed in Table 2, although 2 patients had subsequent studies. All findings were abnormal, demonstrating demyelination, with additional axonal involvement in 3 patients. Nerve conduction study findings were abnormal, ie, slowing or block, in more than 1 nerve in all but 2 studies. Proximal conduction block, defined as the absence of an F wave or prolonged F-wave latency (adjusted for height) that was not explained by distal dispersion or slowing, was present in 8 patients. Distal block, defined as a greatly prolonged distal latency, was found in numerous nerves in 7 patients. Conduction block along the course of a nerve, excluding a proximal or distal block in the same nerve, was demonstrated in 6 patients. Slowing of nerve conduction velocities was present in 6 patients. Reduction in compound muscle action potential amplitudes, not meeting criteria for conduction block, was present in 3 patients and suggested axonal involvement. Sensory nerve action potentials were absent or were reduced in amplitude in 4 patients and, in another, there was a prolonged distal latency. Only 3 of the patients had denervation changes (fibrillations or positive sharp waves) on needle EMG; these studies were performed later in the course of illness (14-21 days).

Six of the patients were available for reevaluation 2 months or longer after the onset of their illness (Table 1). All had normal or near-normal neurologic examination results at that time and those who were employed previously returned to work in 2 to 6 months. Two patients still had sensory symptoms at 6 months, 1 with dysesthesias in the hands and 1 with paresthesias of the hands and feet. There was no apparent difference in the speed of recovery or the residual symptoms between the 8 patients treated with intravenous immunoglobulin or plasmapheresis compared with the 4 who were untreated.

In all likelihood there are mild cases of GBS that never come to the attention of a neurologist. In our series, 4.7% of the patients had persistently mild weakness and were able to walk throughout their illness. This compares with 28% in a study from the Netherlands,² 9% in our previous prospective series of 120 patients with GBS,³ 9% in the French Cooperative Group on Plasma Exchange in Guillain-Barré Syndrome series of 556 patients,⁴ and 12% of 100 patients with GBS in the prospective series of Winer et al.⁵ The mild cases that came to our attention showed no age or sex preponderance and no seasonal grouping, in contrast with some of the findings in the recent Netherlands survey.² In keeping with the stereotypic syndrome of GBS, most of our patients had a preceding infection, all but 2 had paresthesias, many had significant pain of several varieties, but none had autonomic involvement. The significant pain syndrome in most of these patients that was disproportionate to the mild motor involvement was somewhat surprising. The average time to reach a clinical nadir in terms of power was 8 days, and in all cases occurred within 14 days. Most patients had a normal or near-normal CSF protein level and none had electrophysiologic evidence of denervation within 13 days of the onset of weakness. There was no particular pattern of EMG changes that characterized this group except perhaps the absence of substantial axonal findings. The CSF studies and EMG were performed during the early stage of the illness and may have underestimated the peak CSF protein level and the amount of axonal degeneration.

We were unable to determine how many patients with GBS who had mild illness at 8 days, later had a deterioration of their condition, but our impression from the remainder of our case series is that very few cases fell into this category. This average interval of 8 days to reach a nadir was similar to the findings of other studies that included large numbers of patients with GBS of all grades of severity, including our own.³ For example, the nadir was reached within 7 days in 34% (14 days in 70%) of the patients described by Winer et al,⁵ within 7 days in 74% of the patients described by Marshall,⁶ and 12 days was the average in both case series reported by Ravn⁷ and Loffel et al.⁸ Similarly, experience in a retrospective group of 169 patients found that 38% reached nadir by 7 days and 77% within 14 days.³ Furthermore, it has been our experience and that of others that transient improvement in power or paresthesias followed by worsening does not occur except in relation to immune treatment, ie, a rebound. This type of rebound might suggest that the mildest cases should not be treated. In the study by the French Cooperative Group on Plasma Exchange in Guillain-Barré Syndrome, untreated mild cases, which included some patients who would fall in to Medical Research Council grade 3 in addition to grades 1 and 2, had a median time to the onset of motor recovery of 8 days, compared with 4 days in those treated with plasmapheresis. However, all the relapses occurred in the treatment group.⁴ This difference in latency of motor improvement was modest and both their and our data suggest no difference in long-term sequelae between treated and untreated patients with mild affection. Therefore, there seems to be no definitive evidence that treatment is warranted in these patients.

There were no distinguishing clinical or electrophysiologic features between our treated and untreated patients with mild GBS, nor were there important differences between patients with mild GBS and those more severely affected that would allow for the prediction of a persistently mild course. Given that there is no reliable predictive factor that can be obtained from a single visit, it may be advisable to hospitalize these patients for 2 to 3 days to demonstrate a stable trend indicative of a mild course of GBS. Those whose conditions are no worse at approximately 8 days are expected to remain in the mild category. A tentative suggestion might be to wait until the eighth day or so before deciding on treatment of patients who still are able to walk.

Accepted for publication September 6, 2000.

Reprints: Deborah M. Green, MD, Neuroscience Institute, The Queen's Medical Center, 1301 Punchbowl St, Honolulu, HI 96813 (e-mail: dgreen@queens.org).